Pith. sign in

REVIEW 11 cited by

One Diffusion Step to Real-World Super-Resolution via Flow Trajectory Distillation

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2502.01993 v2 pith:SGTJAWB4 submitted 2025-02-04 cs.CV

Jianze Li , Jiezhang Cao , Yong Guo , Wenbo Li , Yulun Zhang This is my paper
classification cs.CV
keywords diffusionmodelone-stepflowmodelsreal-isrimageteacher
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Diffusion models (DMs) have significantly advanced the development of real-world image super-resolution (Real-ISR), but the computational cost of multi-step diffusion models limits their application. One-step diffusion models generate high-quality images in a one sampling step, greatly reducing computational overhead and inference latency. However, most existing one-step diffusion methods are constrained by the performance of the teacher model, where poor teacher performance results in image artifacts. To address this limitation, we propose FluxSR, a novel one-step diffusion Real-ISR technique based on flow matching models. We use the state-of-the-art diffusion model FLUX.1-dev as both the teacher model and the base model. First, we introduce Flow Trajectory Distillation (FTD) to distill a multi-step flow matching model into a one-step Real-ISR. Second, to improve image realism and address high-frequency artifact issues in generated images, we propose TV-LPIPS as a perceptual loss and introduce Attention Diversification Loss (ADL) as a regularization term to reduce token similarity in transformer, thereby eliminating high-frequency artifacts. Comprehensive experiments demonstrate that our method outperforms existing one-step diffusion-based Real-ISR methods. The code and model will be released at https://github.com/JianzeLi-114/FluxSR.

Discussion (0). Sign in to comment.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. OP4KSR: One-Step Patch-Free 4K Super-Resolution with Periodic Artifact Suppression

    cs.CV 2026-05 unverdicted novelty 7.0 of 10

    OP4KSR enables efficient one-step 4K super-resolution without patches by adapting Flux with RoPE rescaling and periodicity loss to suppress artifacts.

  2. FluxFlow: Conservative Flow-Matching for Astronomical Image Super-Resolution

    cs.CV 2026-05 unverdicted novelty 7.0 of 10

    FluxFlow is a conservative pixel-space flow-matching framework for astronomical super-resolution that incorporates real atmospheric uncertainty and a training-free Wiener correction, outperforming baselines on a new 1...

  3. Bridging Restoration and Generation Manifolds in One-Step Diffusion for Real-World Super-Resolution

    cs.CV 2026-04 unverdicted novelty 7.0 of 10

    IDaS-SR achieves one-step real-world super-resolution by bridging restoration and generation manifolds via adaptive inversion noise estimation and continuous trajectory steering.

  4. ScaleResfusion: Residual Rectified Flow based on Residual Vector Field

    cs.CV 2026-07 conditional novelty 6.0 of 10

    ScaleResfusion modifies rectified flow to start from a noisy low-quality image and learn only a residual velocity field, enabling 4-step image restoration with LoRA fine-tuning of billion-scale text-to-image models.

  5. PhysFlow: Frequency Decoupled with Dual-Field Rectified Flow for Remote Photoplethysmography

    cs.CV 2026-06 unverdicted novelty 6.0 of 10

    PhysFlow applies frequency-decoupled dual conditional velocity fields in a rectified flow framework to separately reconstruct trend and amplitude parts of rPPG waveforms for improved robustness.

  6. TUDSR: Twice Upsampling-Diffusion for Higher Super-Resolution

    cs.CV 2026-06 unverdicted novelty 6.0 of 10

    TUDSR applies a twice-upsampling diffusion strategy with chunk-based training to achieve state-of-the-art super-resolution at 1024^2 and 2048^2 resolutions using a one-step GAN on SD2.1-base.

  7. DreamSR: Towards Ultra-High-Resolution Image Super-Resolution via a Receptive-Field Enhanced Diffusion Transformer

    cs.CV 2026-05 unverdicted novelty 6.0 of 10

    DreamSR uses a dual-branch MM-ControlNet with patch-level and global prompts plus a receptive-field enhancement training strategy in a diffusion transformer to reduce over-generation and improve local texture details ...

  8. Bridging Restoration and Generation Manifolds in One-Step Diffusion for Real-World Super-Resolution

    cs.CV 2026-04 unverdicted novelty 6.0 of 10

    IDaS-SR performs one-step real-world super-resolution by predicting severity-aware timesteps to anchor low-quality latents and using continuous trajectory steering to balance structure and texture generation.

  9. TeEFusion: Blending Text Embeddings to Distill Classifier-Free Guidance

    cs.CV 2025-07 conditional novelty 6.0 of 10

    TeEFusion distills classifier-free guidance into text embeddings via linear fusion, enabling a student model to generate images in one forward pass instead of two.

  10. FluxFlow: Conservative Flow-Matching for Astronomical Image Super-Resolution

    cs.CV 2026-05 unverdicted novelty 5.0 of 10

    FluxFlow uses conservative pixel-space flow-matching with uncertainty weights and Wiener test-time correction to outperform baselines on photometric and scientific accuracy for ground-to-space super-resolution, valida...

  11. Bridging Information Asymmetry: A Hierarchical Framework for Deterministic Blind Face Restoration

    cs.CV 2026-01 conditional novelty 5.0 of 10

    Pref-Restore combines AR semantic tokens, a diffusion generator, and DiffusionNFT-style RL to make blind face restoration more consistent, but its deterministic-identity claim is weakened by self-referential rewards a...

Pith tools